T cell acute lymphoblastic leukemia (T-ALL) in children has a dismal overall prognosis including a relapse rate of 25% due to a lack of non-cytotoxic targeted therapies. Relapse is thought to be mediated by leukemia- initiating cells (L-IC) resistant to dexamethasone, an essential component of ALL therapy. Research from my lab and others has shown that T-ALLs are more complex than originally realized, consisting of a mix of genetically related transformed cells, arrested at distinct stages of maturation. Indeed, we have shown that mouse T-ALLs consist of thymic progenitors and differentiated leukemic blasts with distinct self-renewal, proliferative and survival properties. We showed that mouse DN3 stage thymic progenitors are enriched in L- IC that contribute to resistance to ?-secretase inhibitors (GSI) used to inhibit NOTCH in T-ALL. In Aim 1, we will address the complexity of the L-IC and further define its identity by single-cell analyses of the oligoclonal, DN3-enriched L-IC population. To discern the effects of L-IC heterogeneity on dexamethasone responses, we will use a multi-color reporter mouse to monitor effects of dexamethasone treatment on clonal composition and to reveal gene expression changes responsible for dexamethasone resistance in vivo. Recently, we showed that GSI resistance involves epigenetic changes in a rare human T-ALL subclone. We find the RUNX DNA-binding motif enriched in enhancer maps of GSI-resistant human T-ALL cells and resistance is associated with the preferential looping of the MYC promoter to a 3?? enhancer. We demonstrate that Runx1 deletion in mouse T-ALL cells depletes active chromatin marks and decreases MYC expression, resulting in apoptosis. In Aim 2, we will use chromosome conformation capture to determine whether RUNX1 maintains leukemic growth and contributes to GSI resistance by regulating MYC enhancer-promoter interactions. Finally, the knowledge gained in mouse T-ALL models will be translated to human T-ALL by determining the effect(s) of RUNX inhibition on survival and GSI resistance in relapsed pediatric T-ALL patient-derived xenografts. The work proposed in Aim 3 focuses on the novel dexamethasone-resistance genes identified in our siRNA survival-based screen in dexamethasone-sensitive mouse T-ALL cells. In that screen we identified several genes linked to dexamethasone-resistance and/or leukemia suppression in patients, including Ep300, Ikaros and Utx, thereby validating our experimental approach. We will determine how validated genes promote dexamethasone resistance and will examine the ability of these novel genes to accelerate leukemogenesis and alter dexamethasone responses in our mouse and human T-ALL models. Together, the studies proposed in this application will reveal the cellular and molecular (genetic and epigenetic) events responsible for GSI and dexamethasone resistance, setting the stage for pre-clinical studies using inhibitors to the drug-resistance pathways identified herein.